Quality of Life Research

https://doi.org/10.1007/s11136-019-02279-6

Socioeconomic status, social support, oral health beliefs, psychosocial

factors, health behaviours and health‑related quality of life
in adolescents
Andressa Coelho Gomes1 · Maria Augusta Bessa Rebelo1 · Adriana Correa de Queiroz1 ·
Ana Paula Correa de Queiroz Herkrath1 · Fernando José Herkrath2 · Janete Maria Rebelo Vieira1 ·
Juliana Vianna Pereira1 · Mario Vianna Vettore3 

Accepted: 21 August 2019

© Springer Nature Switzerland AG 2019

Abstract
Purpose  This study assessed the relationships between socioecononic status (SES), social support, oral health beliefs, psy-
chosocial factors, health-related behaviours and health-related quality of life (HRQoL) in adolescents.
Methods  A school-based follow-up study involving 376 12-year-old adolescents was conducted in Manaus, Brazil. Baseline
data included sociodemographic characteristics (sex, parental schooling, family income, household overcrowding and number
of goods), social support (SSA questionnaire), oral health beliefs and psychosocial factors (Sense of Coherence [SOC-13
scale] and self-esteem [Rosenberg Self-Esteem Scale]). Health-related behaviours (toothbrushing frequency, sedentary
behaviour, smoking and sugar consumption) and HRQoL [KINDL questionnaire] were assessed at 6-month follow-up.
Structural Equation Modelling assessed the relationships between variables.
Results  Greater social support (β = 0.30), higher SOC (β = 0.23), higher self-esteem (β = 0.23), higher toothbrushing fre-
quency (β = 0.14) and less smoking (β = − 0.14) were directly linked with better HRQoL. SES (β = 0.05), social support
(β = 0.26), oral health beliefs (β = − 0.02) were indirectly linked to HRQoL. Higher SES directly predicted higher toothbrush-
ing frequency (β = 0.14) and less smoking (β = − 0.22). Greater social support also directly predicted higher SOC (β = 0.55),
positive oral health beliefs (β = − 0.31) and higher self-esteem (β = 0.58). Greater social support indirectly predicted less
smoking via oral health beliefs (β = − 0.05) and less sugar consumption via SOC (β = − 0.07).
Conclusion  Socioeconomic status, social support, oral health beliefs and psychosocial factors were important predictors of
adolescent’s health behaviours and HRQoL over 6-month period through direct and indirect mechanisms. Health behaviours
also directly influenced HRQoL.

Keywords  Socioeconomic factors · Psychosocial factors · Health risk behaviour · Quality of life

Introduction

Adolescence is a major transition period in the individual’s

life, and it is characterized by biological, behavioural and
* Mario Vianna Vettore psychological changes that are influenced by social condi-
m.vettore@sheffield.ac.uk tions and family characteristics [1]. Unhealthy behaviours,
such as smoking, alcohol consumption, drug use, unpro-
1
School of Dentistry, Federal University of Amazonas, tected sex, inadequate diet and sedentarism, are commonly
Avenida Ministro Waldemar Pedrosa, 1539, Bairro Praça 14,
Manaus, Amazonas CEP 69.025‑005, Brazil acquired during adolescence and may persist throughout the
2 life-course affecting health and well-being [2]. Adolescence
Fundação Oswaldo Cruz, Instituto Leônidas & Maria Deane,
Rua Terezina, 476, Adrianópolis, Manaus, Amazonas, Brazil is also characterized by the modification and expansion of
3 the social ties of family members, teachers and friends.
Academic Unit of Oral Health, Dentistry and Society, School
of Clinical Dentistry, University of Sheffield, 19 Claremont Social support is strongly related to the emotional and social
Crescent, Sheffield S10 5SX, UK development of adolescents due to the interdependence and

13
Vol.:(0123456789)

influence between individuals interacting with different
social groups. The importance of emotional support, loyalty,
understanding and intimacy on well-being is emphasized
at this stage of life [3–5]. In addition, positive oral health
beliefs, protective psychosocial factors such as high sense
of coherence and greater self-esteem are also considered to
be important aspects that can positively influence adolescent
health behaviour, health-related quality of life and mental
health [6–12].
There is a growing interest in the assessment of health-
related quality of life (HRQoL) in healthcare and health
research. The limitations of normative clinical assessment
of health have resulted in the expansion in the use of HRQoL
measures because the latter considers the subjective aspects
of health as well as how people perceived their own health
and disease [13, 14]. HRQL is a multidimensional construct
defined as “how well a person functions in their life and his
or her perceived well-being in physical, mental, and social
domains of health” [15].
Conceptual frameworks of HRQoL highlight the impor-
tance of socioeconomic position, behaviours and psycho-
social factors on quality of life [16]. Demographic and
socioeconomic characteristics can influence adolescent
health-related behaviours and quality of life [8, 17–20]. Evi-
dence suggests that young people from low socioeconomic
backgrounds are more likely to engage in unhealthy behav-
iours such as tobacco smoking, poor diet and physical inac-
tivity [21]. Socioeconomic status may influence adolescent
health behaviours through psychosocial factors (e.g. coping
mechanisms), oral health beliefs and social ties with family
members and schoolmates [10, 22, 23]. Adolescents with a
high socioeconomic status are more likely to live in intact
families and to develop supportive social networks, which
in turn enhance coping skills to reduce chronic disease risk
behaviours [22, 23]. Furthermore, the influence of socioeco-
nomic status during childhood on adult health behaviours
(e.g. toothbrushing frequency) was mediated by oral health
beliefs during early adulthood [10]. Socioeconomic status
and social networks are significant predictors of adolescent
psychosocial health, for example positive attitudes about
health, greater self-esteem and self-efficacy and member-
ship in peer groups, which in turn promote the adoption
of healthy behaviours [4, 5, 22, 23]. Findings from large
surveys conducted in European countries suggest that higher
parental educational status and family wealth are associated
with better HRQoL of both children and adolescents [24,
25].
Social support may affect health through favouring adap-
tive behaviours in stressful situations [26]. Supportive social
relationships from family members, friends and schoolmates
may positively impact on a range of health-related behav-
iours in adolescents [4, 5, 9, 27, 28]. Data from the BELLA
German study involving children and adolescents revealed
13
Quality of Life Research
that individual factors (e.g. self-efficacy, optimism and sense
of coherence), family and social characteristics (e.g. social
support) are important resources associated with HRQoL
[7]. Direct and indirect mechanisms may explain the influ-
ence of social support and socioeconomic status on health
outcomes and quality of life. Direct pathways suggest that
adolescent health and quality of life can be affected through
material circumstances and psychological mechanisms. The
former include standard of living (e.g. housing conditions
and quality of neighbourhood) and consumption potential
(e.g. financial means to buy healthy food) [29]. Psychologi-
cal mechanisms operate through cognitive and emotional
processes including the development and strengthening
of trust, well-being, self-efficacy, social integration, self-
esteem and mutual cooperation [30, 31]. Indirect effects of
social connections and social status on health result from
behavioural pathways because health-related behaviours in
adolescents are related to social influence from family mem-
bers and friends and mediated by access to material goods
and resources [4, 5, 15, 32, 33].
The possible influence of oral health beliefs and health-
related behaviours on HRQoL among children and adoles-
cents has been investigated [9–12, 34–36]. The findings
suggest a continuity in oral health beliefs about a range of
preventive behaviours and quality of life [10–12]. Positive
oral health beliefs directly predict better overall quality of
life among adolescents in a 6-month follow-up study [12].
Similarly, adolescents with stable favourable oral health
beliefs had a significantly lower prevalence of poor self-
rated oral health during adulthood [11]. Children report-
ing unhealthy behaviours were more likely to have worse
HRQoL even when sex, social context and symptoms were
taken into account [37]. Similarly, data from 12 countries
revealed that better HRQoL was inversely correlated with
health risk behaviours in children and adolescents [38].
Despite the available research on the possible risk factors
of HRQoL in adolescents, the majority of evidence relies on
cross-sectional data and longitudinal studies on this topic are
scarce [9, 34]. Furthermore, the rationale for selecting the
independent variables in the previous studies was not theo-
retically sound. The use of conceptual frameworks is recog-
nized as an important strategy when investigating the predic-
tors and determinants of HRQoL [8, 16, 17]. The predictors
of health and well-being are organized into structural and
intermediary determinants according to the WHO’s concep-
tual framework. The former includes socioeconomic position
(e.g. income and education), while the latter is composed of
psychosocial factors and behavioural characteristics [29].
As far as the authors are aware, no previous studies have
evaluated the simultaneous role of multiple structural and
intermediary determinants of HRQoL in adolescents using
a conceptual framework and a prospective study design. We
hypothesized that adolescents with a low socioeconomic
Quality of Life Research
status, low social support, negative health beliefs and lower
levels of protective psychosocial factors are more likely to
report unhealthy behaviours and poor HRQoL. It was also
hypothesized that unhealthy behaviours would predict poor
HRQoL in adolescents. In addition, we also conjectured that
the influence of socioeconomic status, social support, health
beliefs and protective psychosocial factors on HRQoL was
mediated by health behaviours.
Materials and methods
Study design and participants
A school-based follow-up study involving 12-year-old ado-
lescents was conducted in the eastern region of the city
of Manaus, Brazil. Baseline data included demographics
(sex) and socioeconomic status (parental schooling, fam-
ily income, household overcrowding and number of goods),
social support, oral health beliefs and psychosocial fac-
tors (self-esteem and sense of coherence). Health-related
behaviours (frequency of toothbrushing, sedentary behav-
iour, smoking and sugar consumption) and HRQoL were
assessed at a 6-month follow-up. The eastern region of the
city of Manaus is the second most populous urban area of
the city (population size nearly 450,000 habitants), and the
region with worst social indicators (Gini Index = 0.440).
Adolescence is a transition period from childhood to adult-
hood that typically spans from 12 to 18 years of age. Since
age is a meaningful aspect related to risk-taking behaviours,
psychological changes and HRQoL, this study focused on
Fig. 1  Conceptual framework of the associations between socioeconomic status, sex, social support, oral health beliefs, psychosocial factors,
health behaviours and health-related quality of life (HRQoL)
12 year-old adolescents [39]. Adolescents with any syn-
drome and/or in need of special care, and those using ortho-
dontic appliances, were excluded.
Sampling procedures and study power
A representative sample of 12-year-old adolescents enrolled
in year 7 of public schools in the eastern region of the city
of Manaus was selected using a two-stage sampling process.
Initially, 25 of the 36 schools with students in year 7 were
selected. The selection of schools was proportional to the
number of 12-year-old students in each school. Therefore,
schools with larger number of students were more likely to
be selected. All 12-year-old students from all classes in year
7 of the selected schools were invited to participate.
The final sample size comprised 376 adolescents, consid-
ering a power of 90% to detect statistically significant effects
of 0.20 and 5% statistical significance to estimate a structural
equation analysis model with three latent variables and eight
observed variables [40].
Conceptual framework and measures
The conceptual framework of the present study was adapted
from the WHO conceptual framework of social determinants
of health and well-being [29] (Fig. 1). It was expected a
priori that structural determinants, including poor socio-
economic status (low family income, high household over-
crowding, fewer number of goods and low parental educa-
tion), low social support and sex would predict intermediary
determinants, including negative oral health beliefs, low
13

protective psychosocial factors and unhealthy behaviours.
It was also expected that the aforementioned structural and
intermediary determinants would predict worse HRQoL. In
addition, the intermediary determinants would mediate the
relationship between structural determinants and HRQoL.
Data were collected using self-administered standard-
ized questionnaires and interviews completed by adoles-
cents under the researcher’s supervision in a private room
at each school. The participants’ parents or guardians pro-
vided information concerning socioeconomic status through
a structured questionnaire completed at home. The socio-
economic questionnaire reached the parents or guardians
through the adolescents.
Socioeconomic status
Socioeconomic status was a latent variable measured by
four indicators: family income, household overcrowding,
number of goods and parental schooling. Family income
was categorized as: 1 = up to ½ minimum wage, 2 = > ½
minimum wage up to 1 minimum wage and 3 = > 1 mini-
mum wage. The minimum wage in the study period was
U$ 271.09 [41]. Household overcrowding was assessed by
dividing the number of residents by the number of rooms
in the household and categorized as 1 = 3 or more people
per room, 2 = between 2 and 2.99 people per room, 3 = 0
to 1.99 people per room. Number of goods in the house-
hold was assessed according to a list of 11 durable goods at
home. Parental schooling was registered as the total number
of years of schooling with approval and categorized into 3
groups: 1 = 1–7 years of schooling, 2 = 8–11 years of school-
ing and 3 = ≥ 12 years of schooling.
Social support
Social support was a latent variable using the dimensions
of the Social Support Appraisals (SSA) instrument as indi-
cators [42]. SSA assesses emotional support according to
Cobb’s theoretical definition of social support: beliefs that
one is loved, respected and esteemed by and involved with
family, friends and others [23]. The SSA valid version for
the Brazilian population was used [43]. The questionnaire
comprises a total of 30 items grouped into four dimensions
that assesses emotional social support from family (e.g. My
family cares for me very much), friends (e.g. My friends
look out for me), teachers (e.g. My teachers like me) and
others (e.g. I fell valued by other people). SSA questionnaire
is a 6-point Likert scale with the following response options:
1 = “I fully agree”, 2 = “I strongly agree”, 3 = “I agree a lit-
tle”, 4 = “I disagree somewhat”, 5 = “I strongly disagree”
and 6 = “I fully disagree”. The total score of SSA question-
naire is obtained by adding up the scores of the items and
13
Quality of Life Research
vary from 30 to 180. A higher SSA score indicates greater
social support.
Oral health beliefs
Adolescents completed a questionnaire about oral health
beliefs proposed by Broadbent et al. concerning six behav-
iours: avoiding a large amount of sweet foods, using fluoride
toothpaste, visiting the dentist regularly, keeping teeth and
gums very clean, drinking fluoridated water and floss [11].
Participants ranked each belief as: 1 = ”Extremely impor-
tant”, 2 = ”Reasonably important”, 3 = ”It does not matter
much/Not very important” or 4 = ”Nothing important”. The
oral health beliefs items are combined resulting in a final
score ranging from 6 (very favourable) to 24 (very unfavour-
able). The higher the score, the worse the oral health beliefs.
Psychosocial factors
Psychosocial factors were sense of coherence (SOC) and
self-esteem. Adolescent SOC was assessed using the Brazil-
ian version of the SOC-13 scale [44, 45]. The scores related
to negative SOC items were reversed. The final score of
the SOC measure was obtained by summing the points of
the 13 items, ranging from 13 to 65. The higher score, the
higher the SOC. Self-esteem was assessed through the Bra-
zilian version of the Rosenberg’s Self-Esteem Scale [46, 47].
The 10 items were answered on a four-point Likert scale:
4 = ”I totally agree”, 3 = ”I agree”, 2 = ”I disagree” and 1 = ”I
totally disagree”. The total self-esteem score comprised the
sum of the items and can vary from 10 to 40. The higher
score indicates higher self-esteem. The above-mentioned
psychosocial measures have been previously used to assess
the relationship between psychosocial factors and quality of
life in adolescents [12].
Health‑related behaviours
Smoking, sedentary behaviour and frequency of tooth-
brushing were measured according to questions used in the
National School-Based Health Survey (Pesquisa Nacional
de Saude do Escolar—PeNSE) [48]. Smoking was assessed
according to the question: “During the last 30 days, have you
smoked cigarette?” (1 = Yes, 2 = No). Sedentary behaviour
was measured in response to the question: “In an ordinary
weekday, how many hours per day do you spend in watching
TV, using computer and video games or doing other sitting
activities?” (1 = < 1 h per day, 2 = 1 to 2 h per day, 3 = 3 to
4 h per day or 4 = > 4 h per day). The item “How many times
do you brush your teeth per day?” (1 = up to 2 times a day,
2 = 3 times a day or more) was used to assess the frequency
of toothbrushing.
Quality of Life Research
The annual amount of sugar consumption was evaluated
through a Food Frequency Questionnaire (FFQ) applied
by trained examiners using coloured figures of real-size
food portions and drinks. The FFQ consists of a list of
cariogenic foods and drinks previously used by Peres
et al. [49], and originally advocated by Chaffe et al. [50].
The adolescents were questioned about the frequency and
amount of consumption of fourteen groups of foods and
drinks with cariogenic potential. These groups are based
on their respective amounts of sucrose and/or associations
with dental caries. The daily intake of sugar-related foods
and drinks was transformed into annual amount of sugar
consumption.
Health‑ related quality of life
Health-related quality of life was measured using the Kiddo-
KINDL questionnaire validated for Brazilian adolescents
aged from 12 to 16 years-old [51]. The Kiddo-KINDL was
originally developed by Bullinger (1994) and reviewed by
Ravens-Sieberer and Bullinger [52]. The Kiddo-KINDL
questionnaire contains 24 items distributed into six subscales
corresponding to the following dimensions: physical well-
being, emotional well-being, self-esteem, family, friends and
daily routine (school). The total score of Kiddo-KINDL was
obtained by the sum of all items and can vary from 24 to
120. HRQoL was a latent variable using the scores of each
dimension as indicators. Higher scores of Kiddo-KINDL
indicate better HRQoL.
Pilot study and instrument reliability
A pilot study was conducted involving ten 12-year-old stu-
dents who did not participate in the main study. The students
were interviewed to verify understanding of the items in
the questionnaires. Ten percent of the participants of the
main study were re-interviewed in order to assess the tem-
poral reliability of the instruments. Internal consistency
and reliability of the instruments were evaluated through
Cronbach’s alpha coefficient and Intraclass Correlation
Coefficient (ICC), respectively. Cronbach’s alpha coefficient
greater than 0.80 suggests high reliability, whereas Cron-
bach’s alpha between 0.50 and 0.80, and below than 0.50
indicate moderate and low reliability, respectively [53]. ICCs
between 0.50 and 0.75, between 0.75 and 0.90 and greater
than 0.90 indicate moderate, good and excellent reliability,
respectively [54]. The Cronbach’s alpha coefficients were:
social support = 0.876, self-esteem = 0.844, SOC = 0.674,
QFA = 0.748 and HRQoL = 0.810. The ICCs were: social
support = 0.892, self-esteem = 0.878, SOC = 0.888, oral
health beliefs = 0.701; QFA = 0.720 and HRQoL = 0.885.
Data analysis
Descriptive analysis reported the distribution of the vari-
ables through means and standard deviations (continuous
variables) and proportions (categorical variables). Con-
firmatory factorial analysis (CFA) was used to evaluate the
measurement model involving three latent variables (socio-
economic status, social support and HRQoL). Structural
equation modelling (SEM) examined the direct and indirect
relationships between observed and latent variables accord-
ing to the conceptual framework (Fig. 1) using SPSS AMOS
24.0 software.
The standardized total effects composed of standardized
direct effects (a direct path from one variable to another)
and standardized indirect effects (a pathway mediated by
other variables) were estimated. The 95% confidence inter-
vals (95% CIs) were estimated using the maximum likeli-
hood method via bias-corrected bootstrap to assess whether
mediation was present by testing the statistical significance
of the indirect effects, with 900 resampling from the original
data set in order to derive less biased standard errors [55].
The adequacy of the measurement and structural models was
evaluated according to the following fit indexes and thresh-
old values: χ 2/df < 3.0, standardized root-mean-square
residual (SRMR) ≤ 0.08, comparative fit index (CFI) ≥ 0.90,
goodness of fit index (GFI) ≥ 0.90 and root-mean-square
error of approximation (RMSEA) ≤ 0.06 [56]. The non-
significant direct paths were removed from the full model
to estimate a statistically parsimonious model.
Ethical considerations
The present study was conducted in accordance with the
Declaration of Helsinki and the research protocol was
approved by the Ethics Committee of the Federal Univer-
sity of Amazonas (Protocol no. 57273316.1.0000.5020). All
participants of this study and their parents provided signed
informed consent before data collection.
Results
Figure 2 presents the flow chart for selecting the partici-
pants. Initially 442 adolescents were assessed for participa-
tion according to the eligibility criteria. 27 adolescents were
excluded from the study due to current use of orthodon-
tic appliances. The remaining 415 eligible participants all
agreed to participate. Nine participants were excluded from
the analysis because of incomplete data. The baseline and
follow-up data collection included 406 and 376 participants,
respectively (retention rate of 96.6%). Female adolescents
represented 56.4% of the sample and most of them were
from low-income families. The majority of participants’
13
 Quality of Life Research

Table 1  Socioeconomic status, sex, social support, oral health beliefs,

psychosocial factors, health-related behaviours and health-related
quality of life (HRQoL)
Total
N = 376

Socioeconomic characteristics
 Family income
   ≤ ½ BMW 100 (26.6%)
  ½ to 1 BMW 148 (39.4%)
  > 1 BMW 128 (34.0%)
 N of residents per room
  ≥ 3 35 (9.3%)
  2 to 2.99 73 (19.4%)
  0 to 1.99 268 (71.3%)
 Number of durable goods, mean 6.7 (SD ± 2.5)
 Parental schooling
Fig. 2  Flow chart of participants at baseline and follow-up periods   1 to 7 years 64 (17.0%)
  8 to 11 years 270 (71.8%)
  ≥ 12 years 42 (11.2%)
parents or guardians reported having between 8 and 11 years  Sex
  Female 164 (43.6%)
of schooling (71.8%) and residing predominantly in house-
  Male 212 (56.4%)
holds with less than two residents per room (Table 1).
 Social support
The confirmatory factor analysis (CFA) evaluated the
  Total score, mean (± SD) 141.5 ± 16.9
measurement model for the latent variables; socioeconomic
  Friends, mean (± SD) 33.1 ± 5.7
status, social support and HRQoL (Fig. 3). The items that
  Relatives, mean (± SD) 42.2 ± 5.4
confirmed the latent variable “social support” were SSA
  Teachers, mean (± SD) 29.7 ± 5.2
dimensions perceived support from friends (β = − 0.703),
  Others, mean (± SD) 36.5 ± 5.7
parents (β = − 0.683), teachers (β = − 0.569) and others
 Oral health beliefs, mean (± SD) 8.6 ± 2.5
(β = − 0.836). The items that confirmed the latent variable Psychosocial factors
“socioeconomic status” were family income (β = 0.581),  Sense of coherence, mean (± SD) 44.8 ± 7.0
parental schooling (β = 0.284), household overcrowding  Self-esteem, mean (± SD) 28.6 ± 4.1
(β = − 0.318) and number of goods (β = 0.474). The items Health-related behaviours
that confirmed the latent variable “HRQoL” were the Kiddo-  Toothbrushing frequency
KINDL dimensions physical well-being (β = − 0.498), emo-   ≤ 2 times per day 161 (42.8%)
tional well-being (β = − 0.686), self-esteem (β = − 0.612),   ≥ 3 times per day 215 (57.2%)
family (β = − 0.579), friends (β = − 0.590) and daily routine  Sedentary behaviour
(β = -0.446). The highest values of R2 were 0.836 (others) for   < 1 h per day 123 (32.7%)
social support, 0.666 (emotional well-being) for HRQoL and   1 to 2 h per day 98 (26.1%)
0.581 (family income) for socioeconomic status.   3 to 4 h per day 80 (21.3%)
Structural equation modelling supported the hypoth-   > 4 h per day 75 (19.9%)
esized model (full model) with the following values:  Smoking
SRMR = 0.047, CFI = 0.920, GFI = 0.935, RMSEA = 0.044.   Never smoker 355 (94.4%)
No significant association between sex and other variables   Previous or current smoker 21 (5.6%)
was observed in the full model. Variable sex and non-sig-  Sugar consumption (kg/year), mean (± SD) 7.9 ± 5.6
nificant direct paths were removed to obtain the statistical HRQoL
parsimonious model. The fit indices of the full, measurement  Total score, mean (± SD) 85.3 ± 12.4
and parsimonious models are presented in Table 2.  Physical well-being, mean (± SD) 14.8 ± 2.8
The parsimonious model showed good fit since all criteria  Emotional well-being, mean (± SD) 15.3 ± 3.0
were met. The direct and indirect relationships of this model  Self-esteem, mean (± SD) 12.9 ± 3.8
are summarized in Fig. 4. Greater social support (β = 0.30),  Family, mean (± SD) 15.5 ± 3.2
 Friends, mean (± SD) 13.7 ± 3.3
higher SOC (β = 0.23), higher self-esteem (β = 0.23), fre-
 Daily routine, mean (± SD) 13.1 ± 2.5
quency of toothbrushing (β = 0.14) and lower frequency of
smoking (β = − 0.14) were directly related to better HRQoL. BMW Brazilian minimum wage

13
Quality of Life Research
Better socioeconomic status directly predicted higher
toothbrushing frequency (β = 0.14) and lower smoking
frequency (β = − 0.22). Having greater social support was
a direct predictor for higher SOC (β = 0.55), positive oral
health beliefs (β = − 0.31) and higher self-esteem (β = 0.58).
Significant indirect relationships between socioeconomic
status and HRQoL (β = 0.05), social support (β = 0.26) and
oral health beliefs were identified. In addition, greater social
support was an indirect predictor for lower frequency of
smoking via oral health beliefs (β = − 0.05) and lower sugar
consumption via SOC (β = − 0.07). The parameters of the
direct and indirect effects are described in Fig. 4.
Discussion
The present follow-up study assessed the relationships
between socioeconomic status, social support, health-
related behaviours, oral health beliefs, psychosocial factors
Fig. 3  Likelihood estimates for
the confirmatory factor analysis
(measurement model)
and HRQoL in 12 year-old Brazilian adolescents. Overall,
the findings of this study support the hypothesis that low
socioeconomic status, poor social support, negative oral
health beliefs and lower levels of protective psychosocial
factors were significantly associated with unhealthy behav-
iours and poor HRQoL in adolescents. The hypothesis of
a negative effect of unhealthy behaviours on adolescent
HRQoL was partially confirmed. Similarly, the mediation
effect of health behaviours on the influence of socioeco-
nomic status, social support, oral health beliefs and pro-
tective psychosocial factors on HRQoL was observed to a
certain extent. Therefore, our results support the temporal
relationships between socioeconomic inequalities, social
support, oral health beliefs, psychosocial factors, health
behaviours and HRQoL among adolescents. These find-
ings suggest the importance of intermediary social deter-
minants of adolescent health and well-being. The relation-
ships were tested according to a hypothesized conceptual
framework that included socioeconomic factors, oral
13

Table 2  Fit indices for the confirmatory factor analysis of full, meas-
urement and parsimonious models
Model χ2/df ratio GFI CFI SRMR RMSEA
Full 1.722 0.935 0.920 0.047 0.044
Measurement model 1.522 0.960 0.963 0.042 0.037
Parsimonious 1.623 0.931 0.922 0.048 0.041
Model Full: Theoretical model. Measurement model: confirmatory
factor analysis between latent variables (Socioeconomic status, social
support and health-related quality of life [HRQoL]). Parsimonious
model: Associations between socioeconomic status, social support
psychosocial factors, health-related behaviours and HRQoL with
multiple direct and indirect effects model with pathways between all
adjacent and non-adjacent levels
χ2/df ratio Chi square and degrees of freedom ratio, GFI Goodness
of fit statistics; CFI comparative fit index, SRMR standardized root-
mean-squared residual, RMSEA root-mean-square error of approxima-
tion
health beliefs, psychological characteristics, behaviours
and HRQoL using structural equation modelling, which
is considered to be the most adequate statistical approach
when evaluating multiple relationships between variables.
The use of a representative and random sample suggests
that our findings are potentially applicable to other pop-
ulations with similar demographic and socioeconomic
characteristics.
Socioeconomic status was a relevant determinant of
health behaviours in the studied population because better
socioeconomic status predicted greater frequency of tooth-
brushing and lower smoking. This finding reinforces the
Fig. 4  Parsimonious model of associations between socioeconomic
status, social support, oral health beliefs, psychosocial factors, health
behaviours and health-related quality of life (HRQoL). Direct effects
13
Quality of Life Research
importance of socioeconomic characteristics on oral and
general health behaviours, which is in agreement with previ-
ous research [57, 58]. In addition, frequency of toothbrush-
ing, smoking and oral health beliefs mediated the relation-
ship between socioeconomic status and HRQoL. Although
previous research reported the link between socioeconomic
status and adolescent HRQoL, epidemiological investiga-
tions assessing the possible mediators between socioeco-
nomic factors and HRQoL are scarce [59]. Since our study
simultaneously assessed the role of socioeconomic condi-
tions, oral health beliefs, psychosocial factors and behav-
ioural characteristics on HRQoL, it was possible to reveal
the potential pathways by which low social background
influences HRQoL [16, 24, 25].
Adolescent perception of social support was associ-
ated with oral health beliefs, and all psychosocial factors
included in this study. This result demonstrates the relevance
of supportive social relationships on enhancing adolescent
oral health beliefs, SOC and self-esteem. The link between
social support and psychosocial factors suggests that adoles-
cents perceiving more support from their social ties tend to
develop a greater sense of confidence, belonging and care.
In addition, social support was indirectly linked to smoking
mediated by oral health beliefs. This finding emphasizes the
protective effect of social support on unhealthy behaviours
in adolescents, including less psychological complaints [4,
5]. It must be noted that social support was also a direct
predictor of HRQoL, revealing the importance of social ties
on quality of life and well-being of adolescents, which tends
to remain over time [4, 5, 25].
are indicated by solid lines. Indirect effects are indicated by dashed
lines. *P < 0.05, **P < 0.01
Quality of Life Research
Smoking and toothbrushing were meaningful health-
related behaviours associated with HRQoL among adoles-
cents. The association between smoking and poor quality of
life has been widely studied among adults [60]. However,
studies involving adolescents are scarce [20]. In addition,
few studies have explored the relationship between tooth-
brushing and HRQoL since previous evidence relies on the
link between toothbrushing and OHRQoL [61, 62]. The
lack of association between sedentarism and HRQoL was
an unexpected finding. The use of a single item to assess
adolescent sedentary behaviour might be the possible expla-
nation for this result.
Psychosocial factors investigated in the present study
(SOC and self-esteem) directly predicted adolescent
HRQoL. Furthermore, oral health beliefs were indirectly
related to adolescent HRQoL. The association of SOC and
self-esteem with quality of life among adolescents has been
reported and our findings confirm previous results [6–8, 17].
However, few studies have identified the association between
oral health beliefs and HRQoL. The indirect effect between
these two variables was mediated by smoking. Longitudinal
studies using the life-course approach suggested that oral
health beliefs are related to dental caries, oral hygiene and
self-assessment of oral health [50, 62]. A substantial pro-
portion of the population tends to change their beliefs about
health practices between adolescence and young adulthood
[11]. Therefore, the link between oral health beliefs, smok-
ing and HRQoL in adolescents is a potential topic for future
intervention studies. Our findings emphasize the importance
of oral health beliefs and psychosocial factors on adolescent
health and well-being. Along with health-related behaviours,
oral health beliefs and psychosocial factors may be consid-
ered modifiable risk factors in adolescents. The identifica-
tion of those risk factors provides multiple opportunities to
improve adolescent quality of life through tackling social
inequalities in health using inter-sectoral approaches [59].
The present study benefits from the advantages of using
SEM that simultaneously analyses the direct and indi-
rect complex relationships according to a pre-established
conceptual framework [29]. In this study, HRQoL, social
support and socioeconomic status were analysed as latent
variables to represent multidimensional measures. Using
family income, household overcrowding, number of goods
and parental schooling as indicators of socioeconomic sta-
tus overcame the well-known limitation of measuring social
status through a single characteristic. The use of latent
variables to represent HRQoL and social support using the
instrument’s dimensions provides more reliable data and
less measurement errors. Previous research has employed
SEM to investigate biological, functional, socioeconomic
and psychological determinants of HRQoL in adolescents [8,
9, 34]. However, the cross-sectional nature of these studies
imposes limitations concerning the temporal relationships
between variables.
Some limitations of this study need to be considered. The
studied sample is a specific school cohort of adolescents
attending public schools and living with ​​social disadvan-
tage. Consequently, our findings should not be extrapolated
to other age groups and populations from different socioeco-
nomic backgrounds. Although previous research suggests
the importance of health conditions on HRQoL, clinical
variables were out of the scope of this study and therefore
were not examined. It has been suggested that the predictors
of adolescent quality of life should be investigated according
to sex. This approach was not adopted in this study because
it would have significantly reduced the sample size and
therefore the power of the study.
Conclusion
The present study elucidated the complex relationships
between socioeconomic status, social support, oral health
beliefs, psychosocial factors, health behaviours and HRQoL
among adolescents. Socioeconomic status, social support,
oral health beliefs and psychosocial factors were important
predictors of health behaviours and HRQoL through direct
and indirect mechanisms. In addition, health behaviours
directly influenced HRQoL. Our findings suggest the need
to develop and test interventions addressing multiple modi-
fiable behavioural and psychosocial risk factors to improve
adolescent quality of life. The study highlights the impor-
tance of taking into account multiple aspects of adolescent
life on the development of health promotion strategies and
healthcare services planning in order to enhance quality
of life. A comprehensive approach to improve adolescents
health and quality of life should consider socioeconomic
status, psychosocial characteristics, health beliefs, level of
perceived social support and behaviours. In addition, health
care services should move towards a more patient-centred
approach since individual characteristics were meaningful
aspects of adolescent quality of life. Finally, inter-sectoral
policies to reduce social inequalities can possibly improve
adolescent behaviours, health beliefs and psychosocial fac-
tors, and therefore would positively impact on their health
and well-being.
Acknowledgements  The authors thank Coordination for the Improve-
ment of Higher Education Personnel (CAPES), Ministry of Education,
Brazil for the financial support to the Postgraduate Programme in Den-
tistry at the Federal University of Amazonas.
Funding  This study was funded by the National Council for Scientific
and Technological Development—CNPq, Brazil, research Grant No.
423309/2016.
13

Compliance with ethical standards  12. Baker, S. R., Mat, A., & Robinson, P. G. (2010). What psychoso-
Conflict of interest  All authors declare that they have no conflicts of
interest.
Ethical approval  The study was approved by the Research Ethics
Committee of the Federal University of Amazonas under Protocol No.
57273316.1.0000.5020. All procedures performed in studies involving
human participants were in accordance with the ethical standards of
the institutional and/or national research committee and with the 1964
Helsinki declaration and its later amendments or comparable ethical
standards.
Informed consent  Informed consent was obtained from all individual
participants included in the study.
References
1. Sawyer, S. M., Afifi, R. A., Bearinger, L. H., Blakemore, S. J.,
Dick, B., Ezeh, A. C., et al. (2012). Adolescence: A foundation
for future health. The Lancet, 379(9826), 1630–1640.
2. Jackson, C. A., Henderson, M., Frank, J. W., & Haw, S. J. (2012).
An overview of prevention of multiple risk behaviour in adoles-
cence and young adulthood. Journal of Public Health, 34(suppl.
1), i31–i40.
3. Chu, P. S., Saucier, D. A., & Hafner, E. (2010). Meta-analysis of
the relationships between social support and well-being in chil-
dren and adolescents. Journal of Social and Clinical Psychology,
29(6), 624–645.
4. Zambon, A., Morgan, A., Vereecken, C., Colombini, S., Boyce,
W., Mazur, J., et al. (2010). The contribution of club participa-
tion to adolescent health: Evidence from six countries. Journal of
Epidemiology and Community Health, 64(01), 89–95.
5. Springer, A., Parcel, G., Baumler, E., & Ross, M. (2006). Sup-
portive social relationships and adolescent health risk behavior
among secondary school students in El Salvador. Social Science
and Medicine, 62(7), 1628–1640.
6. Länsimies, H., Pietilä, A. M., Hietasola-Husu, S., & Kangasniemi,
M. (2017). A systematic review of adolescents’ sense of coher-
ence and health. Scandinavian Journal of Caring Sciences, 31(4),
651–661.
7. Erhart, M., Wille, N., & Ravens-Sieberer, U. (2008). Empower-
ment of children and adolescents-the role of personal and social
resources and personal autonomy for subjective health. Gesund-
heitswesen (Bundesverband der Arzte des Offentlichen Gesund-
heitsdienstes (Germany), 70(12), 721–729.
8. Gaspar, T., Ribeiro, J. P., de Matos, M. G., Leal, I., & Ferreira, A.
(2012). Health-related quality of life in children and adolescents:
Subjective well being. The Spanish Journal of Psychology, 15(1),
177–186.
9. Mohamadian, H., Eftekhar, H., Rahimi, A., Mohamad, H. T., Sho-
jaiezade, D., & Montazeri, A. (2011). Predicting health-related
quality of life by using a health promotion model among Iranian
adolescent girls: A structural equation modeling approach. Nurs-
ing & Health Sciences, 13(2), 141–148.
10. Broadbent, J. M., Zeng, J., Foster Page, L. A., Baker, S. R., Ram-
rakha, S., & Thomson, W. M. (2016). Oral health-related beliefs,
behaviours, and outcomes through the life course. Journal of Den-
tal Research, 95(7), 808–813.
11. Broadbent, J. M., Thomson, W. M., & Poulton, R. (2006). Oral
health beliefs in adolescence and oral health in young adulthood.
Journal of Dental Research, 85(4), 339–343.
13
Quality of Life Research
cial factors influence adolescents’ oral health? Journal of Dental
Research, 89(11), 1230–1235.
13. Gill, T. M., & Feinstein, A. R. (1994). A critical appraisal of
the quality of quality-of-life measurements. Journal of American
Medical Association, 272(8), 619–626.
14. Guyatt, G. H., & Cook, D. J. (1994). Health status, quality of
life, and the individual. Journal of American Medical Association,
272(8), 630–631.
15. Hays, R. D., & Reeve, B. B. (2010). Measurement and modeling
of health-related quality of life. In J. Killewo, H. K. Heggenhou-
gen, & S. R. Quah (Eds.), Epidemiology and demography in pub-
lic health (pp. 195–205). San Diego: Academic Press.
16. Bakas, T., McLennon, S. M., Carpenter, J. S., Buelow, J. M., Otte,
J. L., Kathleen, M., et al. (2012). Systematic review of health-
related quality of life models. Health and Quality of Life Out-
comes, 10(1), 134.
17. Otto, C., Haller, A. C., Klasen, F., Hölling, H., Bullinger, M.,
Ravens-Sieberer, U., et al. (2017). Risk and protective factors of
health-related quality of life in children and adolescents: Results
of the longitudinal BELLA study. PLoS ONE, 12(12), e0190363.
18. Riesch, S. K., Kedrowski, K., Brown, R. L., Temkin, B. M., Wang,
K., Henriques, J., et al. (2013). Health-risk behaviors among a
sample of US pre-adolescents: Types, frequency, and predic-
tive factors. International Journal of Nursing Studies, 50(8),
1067–1079.
19. Michel, G., Bisegger, C., Fuhr, D. C., & Abel, T. (2009). Age
and gender differences in health-related quality of life of children
and adolescents in Europe: A multilevel analysis. Quality of Life
Research, 18(9), 1147.
20. Dube, S. R., Thompson, W., Homa, D. M., & Zack, M. M. (2012).
Smoking and health-related quality of life among US adolescents.
Nicotine & Tobacco Research, 15(2), 492–500.
21. Hanson, M. D., & Chen, E. (2007). Socioeconomic status and
health behaviors in adolescence: A review of the literature. Jour-
nal of Behavioral Medicine, 30(3), 263.
22. Lowry, R., Kann, L., Collins, J. L., & Kolbe, L. J. (1996). The
effect of socioeconomic status on chronic disease risk behaviors
among US adolescents. Journal of American Medical Association,
276(10), 792–797.
23. Sajjadi, H., Jorjoran Shushtari, Z., Mahboubi, S., Rafiey, H., &
Salimi, Y. (2018). Effect of socio-economic status, family smok-
ing and mental health through social network on the substance use
potential in adolescents: A mediation analysis. Public Health, 157,
14–19.
24. Von Rueden, U., Gosch, A., Rajmil, L., Bisegger, C., & Ravens-
Sieberer, U. (2006). Socioeconomic determinants of health related
quality of life in childhood and adolescence: Results from a Euro-
pean study. Journal of Epidemiology and Community Health,
60(2), 130–135.
25. Erhart, M., Ottova, V., Gaspar, T., Jericek, H., Schnohr, C.,
Alikasifoglu, M., et al. (2009). Measuring mental health and
well-being of school-children in 15 European countries using the
KIDSCREEN-10 Index. International Journal of Public Health,
54(2), 160–166.
26. Cobb, S. (1976). Social support as a moderator of life stress. Psy-
chosomatic Medicine, 38(5), 300–314.
27. Draper, C. E., Grobler, L., Micklesfield, L. K., & Norris, S. A.
(2015). Impact of social norms and social support on diet, physi-
cal activity and sedentary behaviour of adolescents: A scoping
review. Child: Care, Health and Development, 41(5), 654–667.
28. Viner, R. M., Ozer, E. M., Denny, S., Marmot, M., Resnick, M.,
Fatusi, A., et al. (2012). Adolescence and the social determinants
of health. The Lancet, 379(9826), 1641–1652.
29. WHO. (2010). A conceptual framework for action on the social
determinants of health. Geneva: World Health Organization.
Quality of Life Research
30. Umberson, D. (1987). Family status and health behaviors: Social
control as a dimension of social integration. Journal of Health and
Social Behavior, 28(3), 306–319.
31. Phongsavan, P., Chey, T., Bauman, A., Brooks, R., & Silove, D.
(2006). Social capital, socio-economic status and psychological
distress among Australian adults. Social Science and Medicine,
63(10), 2546–2561.
32. Sisson, K. L. (2007). Theoretical explanations for social inequali-
ties in oral health. Community Dentistry and Oral Epidemiology,
35(2), 81–88.
33. Berkman, L. S., & Glass, T. (2006). Social integration, social net-
works, social support, and health. In M. Marmot & R. G. Wilkin-
son (Eds.), Social determinants of health (pp. 137–173). London:
Oxford University Press.
34. Franco-Paredes, K., Díaz-Reséndiz, F. J., Hidalgo-Rasmussen, C.
A., & Bosques-Brugada, L. E. (2019). Health-related quality-of-
life model in adolescents with different body composition. Eating
and Weight Disorders-Studies on Anorexia, Bulimia and Obesity,
24(1), 143–150.
35. Wu, X. Y., Han, L. H., Zhang, J. H., Luo, S., Hu, J. W., & Sun, K.
(2017). The influence of physical activity, sedentary behavior on
health-related quality of life among the general population of chil-
dren and adolescents: A systematic review. PLoS ONE, 12(11),
e0187668.
36. Muros, J. J., Salvador Pérez, F., Zurita Ortega, F., Gámez Sánchez,
V. M., & Knox, E. (2017). The association between healthy life-
style behaviors and health-related quality of life among adoles-
cents. Jornal de Pediatria, 93(4), 406–412.
37. Chen, X., Sekine, M., Hamanishi, S., Wang, H., Gaina, A.,
Yamagami, T., et al. (2005). Lifestyles and health-related qual-
ity of life in Japanese school children: A cross-sectional study.
Preventive Medicine, 40(6), 668–678.
38. Dumuid, D., Olds, T., Lewis, L. K., Martin-Fernández, J. A., Katz-
marzyk, P. T., Barreira, T., et al. (2017). Health-related quality of
life and lifestyle behavior clusters in school-aged children from
12 countries. The Journal of Pediatrics, 183, 178–183.
39. Jaworska, N., & MacQueen, G. (2015). Adolescence as a unique
developmental period. Journal of Psychiatry and Neuroscience,
40(5), 291–293.
40. Westland, J. C. (2010). Lower bounds on sample size in structural
equation modeling. Electronic Commerce Research and Applica-
tions, 9(6), 476–487.
41. Exchange rates. World Currency Exchange Rates and Currency
Exchange Rate History. https​://www.excha​nge-rates​.org/Rate/
USD/BRL/8-31-2016. Accessed 11 June 2018.
42. Vaux, A., Phillips, J., Holly, L., Thomson, B., Williams, D., &
Stewart, D. (1986). The social support appraisals (SS-A) scale:
Studies of reliability and validity. American Journal of Commu-
nity Psychology, 14(2), 195–218.
43. Squassoni, C. E., & Matsukura, T. S. (2014). Adaptacao transcul-
tural da versao Portuguesa do social support appraisals para o
Brasil. Psicologia: Reflexao & Critica, 27(1), 71–80.
44. Antonovsky, A. (1987). Unraveling the mystery of health: How
people manage stress and stay well. San Francisco: Jossey-bass.
45. Bonanato, K., Branco, D. B. T., Mota, J. P. T., Ramos-Jorge, M.
L., Paiva, S. M., Pordeus, I. A., et al. (2009). Trans-cultural adap-
tation and psychometric properties of the Sense of Coherence
Scale’in mothers of preschool children. Interamerican Journal of
Psychology, 43(1), 144–153.
46. Rosenberg, M. (1989). Society and the adolescent self-image.
Middletown: Wesleyan University Press.
47. Hutz, C. S., & Zanon, C. (2011). Revisão da adaptação, validação
e normatização da Escala de Autoestima de Rosenberg. Avaliacao
Psicologica, 10(1), 41–49.
48. IBGE (2013) Pesquisa Nacional de Saude Escolar-PeNSE. Banco
de dados agregados. Rio de Janeiro: Instituto Brasileiro de Geo-
grafia and Estatistica. http://bibli​oteca​.ibge.gov.br/visua​lizac​ao/
livro​s/liv64​436.pdf. Accessed 9 July 2017.
49. Peres, M. A., Sheiham, A., Liu, P., Demarco, F. F., Silva, A. E. R.,
Assunção, M. C., et al. (2016). Sugar consumption and changes
in dental caries from childhood to adolescence. Journal of Dental
Research, 95(4), 388–394.
50. Chaffee, B. W., Feldens, C. A., Rodrigues, P. H., & Vítolo, M.
R. (2015). Feeding practices in infancy associated with caries
incidence in early childhood. Community Dentistry and Oral Epi-
demiology, 43(4), 338–348.
51. Teixeira, I. P., Novais, I. D. P., Pinto, R. D. M. C., & Cheik, N. C.
(2012). Cultural adaptation and validation of the KINDL question-
naire in Brazil for adolescents between 12 and 16 years of age.
Revista Brasileira de Epidemiologia, 15(4), 845–857.
52. Ravens-Sieberer, U., & Bullinger, M. (1998). News from the
KINDL-questionnaire: A new version for adolescents. Quality of
Life Research, 7, 653.
53. Tan, S. (2009). Misuses of KR-20 and Cronbach’s alpha reliability
coefficients. Education and Science, 34(152), 101–112.
54. Koo, T. K., & Li, M. Y. (2016). A guideline of selecting and
reporting intraclass correlation coefficients for reliability research.
Journal of Chiropractice Medicine, 15(2), 155–163.
55. MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G.,
& Sheets, V. A. (2002). Comparison of methods to test mediation
and other intervening variable effects. Psychological Methods,
7(1), 83–104.
56. Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes
in covariance structure analysis: Conventional criteria versus new
alternatives. Structural Equation Modeling, 6, 1–55.
57. Park, Y. D., Patton, L. L., & Kim, H. Y. (2010). Clustering of
oral and general health risk behaviors in Korean adolescents: A
national representative sample. Journal of Adolescent Health,
47(3), 277–281.
58. Vettore, M. V., Moysés, S. J., Sardinha, L. M. V., & Iser, B. P.
M. (2012). Condição socioeconômica, frequência de escovação
dentária e comportamentos em saúde em adolescentes brasileiros:
Uma análise a partir da Pesquisa Nacional de Saúde do Escolar
(PeNSE). Cadernos de Saúde Pública, 28, s101–s113.
59. Pikhart, H., & Pikhartova, J. (2015). The relationship between
psychosocial risk factors and health outcomes of chronic diseases:
A review of the evidence for cancer and cardiovascular diseases.
WHO Regional Office for Europe.
60. Tian, J., Venn, A. J., Blizzard, L., Patton, G. C., Dwyer, T., & Gall,
S. L. (2016). Smoking status and health-related quality of life:
A longitudinal study in young adults. Quality of Life Research,
25(3), 669–685.
61. Mbawalla, H. S., Masalu, J. R., & Åstrøm, A. N. (2010). Socio-
demographic and behavioural correlates of oral hygiene status
and oral health related quality of life, the Limpopo-Arusha school
health project (LASH): A cross-sectional study. BMC Pediatrics,
10(1), 1–10.
62. Broadbent, J. M., Zeng, J., Foster Page, L. A., Baker, S. R., Ram-
rakha, S., & Thomson, W. M. (2016). Oral health-related beliefs,
behaviors, and outcomes through the life course. Journal of Den-
tal Research, 95(7), 808–813.
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
13